Cindy Morgan reminded the IAB to make hotel reservations for the IAB retreat by 1500 UTC on 1 March 2018.
Joe Hildebrand reminded the IAB that the nominations deadline for the 2018-2019 IAB Chair is 2 March 2018.
The minutes of the 7 February 2018 and the 14 February 2018 meetings were approved.
The internal action item list was reviewed.
Brian Trammell reported that the Plenary Planning Program will meet twice at IETF 101; once for a dry run of the IETF 101 Technical Plenary, and once to discuss future plenaries. Brian will send the description of the IETF 101 Technical Plenary to Alissa Cooper for inclusion in the plenary agenda.
The IAB discussed BOF coverage for IETF 101:
Ted Hardie asked if there was going to be a side meeting on Web Packaging. Mark Nottingham volunteered to find out and report back.
The IAB discussed their agenda for IETF 101. Cindy Morgan will update the IAB’s agenda based on that discussion.
Suzanne Woolf will invite Joe Hall to the Monday breakfast meeting to discuss the social and policy environment of the current Internet architecture.
Brian Trammell will follow up with the MAPRG chairs about reviewing their Research Group during the Thursday breakfast meeting.
Allison Mankin reported that the call for papers for ANRW has been
posted and asked the IAB to forward it along to interested parties.
The IAB briefly discussed Jari Arkko’s draft text for a blog post on consolidation. Jari will send the updated text out to the IAB for review; if no objections are raised by the end of the week, he will post the blog entry.
The IAB approved sending draft-iab-marnew-report-01 out for community review. Cindy Morgan will send the community review message after the IESG approves draft-mm-wg-effect-encrypt,
The IAB approved the response to the liaison on IPv6 work items. Lee Howard will work with Scott Mansfield and Cindy Morgan to get the liaison statement posted in the appropriate places.
Steve Bellovin joined the IAB to give a tech chat on “Censorship, Freedom of Speech, and Architecture.”
The slides are available at: <https://www.cs.columbia.edu/~smb/talks/Internet-censorship.pdf>
Steve Bellovin observed that there are two types of censorship we see on the Internet today: blocking information entirely, and drowning it out. Governments and corporations are both implicated in this censorship.
Governments can block access to certain destinations, look for and delete undesirable content, and use social and regulator pressure to discourage creation of undesirable content.
Corporations can control access to content, which is exacerbated when there is a limited choice of ISPs. Creating interesting content takes resources, and so there is a feedback loop: the bigger a corporations, the better content they can create, which attracts more people and makes the corporation even bigger. Corporations profit when people spend time on their sites, and so the expend effort to keep people there (which means they don’t have time to look at other content).
Steve Bellovin said that to solve censorship, we need:
Steve Bellovin suggested that in the short term, we shouldP
Potential longer term solutions include:
Some of this technology exists, but other items don’t exist yet or are too expensive to deploy. There may also be legal obstacles in certain countries.
Martin Thomson asked about the underlying premise that avoiding censorship is a natural good. Steve Bellovin replied that different cultures view things differently and make different tradeoffs.
Jari Arkko asked what happens when sources are biased. Steve Bellovin replied that corporations are an issue because they are the source of most content. The Internet has democratized access to minor voices, but as things become more concentrated, the minor voices get harder to find.
Erik Nordmark pointed out that there is a societal tension in that we want free speech and anonymous free speech, but we also want accountability. Steve Bellovin replied that studies actually do not support the notion that people behave better on non-anonymous networks. In small communities, social pressure works, but in large ones it does not.